'ContinuousFactor' object has no attribute 'cdf'

[HitosVilla]

When I try to run this code:

std_normal_pdf = lambda x : np.exp(-xx/2) / (np.sqrt(2np.pi))
std_normal = ContinuousFactor(['x'],std_normal_pdf)
std_normal.discretize(RoundingDiscretizer, low=-3, high=3, cardinality=12)

I obtain the following error:

---

AttributeError Traceback (most recent call last)
in
1 std_normal_pdf = lambda x : np.exp(-xx/2) / (np.sqrt(2np.pi))
2 std_normal = ContinuousFactor(['x'],std_normal_pdf)
----> 3 std_normal.discretize(RoundingDiscretizer, low=-3, high=3, cardinality=12)

~\AppData\Local\Continuum\anaconda3\lib\site-packages\pgmpy\factors\continuous\ContinuousFactor.py in discretize(self, method, *args, **kwargs)
178 # TODO: finish this
179 """
--> 180 return method(self, *args, **kwargs).get_discrete_values()
181
182 def reduce(self, values, inplace=True):

~\AppData\Local\Continuum\anaconda3\lib\site-packages\pgmpy\factors\continuous\discretize.py in get_discrete_values(self)
131
132 # for x=[low]
--> 133 discrete_values = [self.factor.cdf(self.low + step/2) - self.factor.cdf(self.low)]
134
135 # for x=[low+step, low+2*step, ........., high-step]

AttributeError: 'ContinuousFactor' object has no attribute 'cdf'

Thank you

[ankurankan]

@HitosVilla Is this somewhere in the examples? Because we currently have very limited support for continuous variables.